Tube Amp, What Dark Sorcery Is This?
Martin_H
Posts: 4,051
He spends most of the time working on the case, but when he builds the amp there's nary a tidbit of info about how it works.
Which made me realize that I don't understand how tubes (thermionic valves in some locales) work? What voltage do they work at? Why do they have so many pins? They're like transistors only weird.
Which made me realize that I don't understand how tubes (thermionic valves in some locales) work? What voltage do they work at? Why do they have so many pins? They're like transistors only weird.
Comments
First the video:
Wow. Totally amazing construction. Incredible attention to detail. Everything as perfect as can be. I love the mash-up of old school woodwork and modern day carbon fibre panel. Did you see how perfect those dove tail joints were! Awesome.
Terrible sound at the end. It's a frikken guitar amp. You know, the most famous guitar amp in the world is the Marshal. Built in a back street workshop in London down to the lowest possible price. Distorted like hell. Guitar players just happened to love that!
Valves:
Nothing to say there. Everything you need to know is a google away. But God bless Sir Owen Willans Richardson https://en.wikipedia.org/wiki/Owen_Willans_Richardson
Why do I love tubes?
1) Usually you want high voltages to get them working. That means danger
2) They need a heater. That means the cat loves to sit on top off the box, whatever it is, in the winter.
3) They have this relaxing glow when in use.
4) Rugged. You are unlikely to blow a tube during your experiments.
5) Simple. So simple to understand. You can almost see your circuit working.
6) Still the most linear (least distorting) characteristics of any amplifying device ever created (Without feedback).
Contrary to MikeDYur I don't believe power and dynamic range are the advantages of tube type amplifiers.
Firstly they are very inefficient so if you want a high power amplifier then solid state is a better bet and now a days Class D. Basically a switch mode power supply that can change its output voltage at audio frequencies.
My last amplifier build had triode amplification and MOSFET power output.
Of course there are still high power RF transmitters built with tubes.
Dynamic range is all about signal to noise ratio. I'm not sure tubes win there.
How do they work?
Hot metals "boil" off electrons. Normally those electrons just flop back down to the metal surface to keep the electric charge of the world in balance.
But, put a high positive voltage nearby and those boiled off electrons will continue flying toward that positive potential, attracted by the electric filed.
Great now we have a circuit, hot negative, cold positive, electrons running around due to the applied electric potential. With the magic property that if you reverse the positive and negative connections no current will flow. Those hot electrons stay stuck to the positive potential of the hot plate they came from. And no electrons are being boiled off from the cold positive end. BINGO we have a diode!
But what about the hot negative, cold positive, with current flowing situation? If we put some little negative charge between the hot negative and cold positive that will tend to repel electrons and less of them will make the trip across. Less current flows. BINGO, we have a way to control the current flow.
So there we have it. Hot cathode, cold anode, a control "grid" in the middle.
Not forgetting you need a heater to make all this work
Why so many pins?
Not really. As we see above you need an anode, cathode and grid connection. Same like collector, emitter, base in a transistor. Or more like drain, gate, source, as tubes are more like FETs in operation.
Often there will be two triodes in a tube so that is 6 pins.
Then you need 2 or 3 pins for the heater(s). That makes 9
We will not go to pentodes and tetrodes here.
Do visit Max Robinson's "Fun With Tubes" pages for everything you need to know about tubes. Without all the audiofool nonsense http://www.funwithtubes.net/
Certainly the characteristics of tubes and transistors are different. But properly engineered audio amplifier has distortion levels so low that I refuse to believe anyone can tell what technology it is using.
When I read about double blind tests where people could not tell the difference between a high end tube amp, a high end transistor amp and a string of crappy old 741 op amps I become convinced it's all phooey.
Me, I reckon the speakers make an order of magnitude more difference to the sound quality.
It's a different story if we are talking guitar amps. Those guys don't want faithful reproduction they want distortion.
Yeah, just searched for the article about a double blind test where a bunch of seasoned session guitarists had to identify the true VOX AC-30 amongst seven different digital modelers. The modelers came out on top 70% of the time. Haven't been able to locate the article.
I am a guitarist of 41 years and have had all the big name stuff. Today I use modelers and can get some great sounds that are stored as presets so that I don't need to keep tweaking knobs on stage. I like one of my modelers so much that I bought SIX backup units...LOL (just preparing for the world tour, you understand).
A musical instrument, and I believe effects peddles and such are part of the instrument, is a very tightly coupled man-machine interface. So many other details are important. The size, the shape, weight, balance, touch, tactile feed back.....
I once met a drum player who expounded on how wonderful his drum peddle was for about an hour. He went into every minute detail. The construction, the action, the balance, everything. He loved that drum peddle as much as his foot that operated it!
I could see it was a wonderfully engineered thing, even if I had no idea what to do with it:)
... and then there's.... "Oxygen free copper"!!!!
Dave
Of course! For better "definition", you absolutely MUST include the $2,000+ Power Cord
Ever felt like you were in the wrong business?
This guy has an opinion, for some people tubes are just compfortable.
http://www.theaudioarchive.com/TAA_Resources_Tubes_versus_Solid_State.htm
£50,000 and not a tube to be found.
"Exakt builds on Linn DS to radically improve the performance of music systems by pushing the lossless digital path all the way to the speaker, keeping the signal digital and only converting to analogue at the last possible stage."
Which is odd because pretty much everything has Class D amplifiers now a days. The final filter components and the speaker are the only analogue parts left.
Reminds me. I have a dev board with an MPS Class D chip on it. Keep meaning to try it out.
Anyway, it's a scientifically proven fact that expensive things sound better than cheap things. The more you pay the better it sounds. Especially if your friends can see how much you spent when they come to visit.
LOL, I love messing with my purist fellow guitarists;
Last year I came across this Peavey Auto-tune guitar, on offer for £145. The DSP is built in to the guitar body and the electronics basically sense which string & fret is being played and generate the appropriate notes. This is the exact same technology used by these zero-talent vocalists that we see on these so-called talent shows. You could make a strangled cat sound like ol' blue eyes.
For giggles, I ordered one, expecting it to be a piece of junk. Boy, was I amazed. I am a bit of a wannabe luthier and was expecting to have make some adjustments to make this thing playable; not necessary. String tension/tuning on this instrument is unimportant and furthermore it's the ONLY guitar on the planet that has perfect intonation. This means that whatever note generated by an open string will be EXACTLY the same on the 12th fret but one octave higher. It doesn't matter what guitar you buy or how much you pay your luthier, you can't achieve this on a conventional instrument. I repeat, this cost £145 (~US$220).
But the really cool thing about this instrument is that one can seamlessly switch from conventional to alternate tunings which is where I mess with my fellow guitarist's heads. I can change key by using the "virtual capo" without having to move away from the first fret, use an open tuning so that the entire chord is played with one finger so that I can do fancy hammer-ons, etc with the other free fingers.
They have no idea that I have this technology because I rubbed off the "Peavey AT-200" logo and with the guitar being plain black, it was easy to refinish. I just tell them it's a cheap self-assembled knock-about guitar because I don't want to risk bringing my nice guitars to the pub.
For these brainwashed purists, technology is "not cool,dude" so there evidently wasn't a huge demand for a self-tuning guitar. I made sure I picked up a backup in case they disappear.
Now, if I only understood the rest of what you were talking about
I see a lot of opinions in that posting but not a single measurement to back them up. Also have to wonder what year that was done. Semiconductors and circuit design have been improving constantly.
However, it is apparently not a tube vs transistor thing. Instead it's a result of the amplifier circuit topology used.
See the section on Harmonic Content and Distortion.
https://en.wikipedia.org/wiki/Tube_sound
It seems I have been reading these arguments since 1970 something.
In the middle of it all you find the occasional double blind test where the results show people cannot reliably tell the difference.
Rod Eliot has some wise observations on all of this : http://sound.westhost.com/valves/valve-trans.html
One thing I do know. If you spend a lot of time and effort building a tube amp, even a small simple one. Draw blood hacking out the aluminium chassis. Burn fingers soldering it up. Electrocute yourself a few times whilst testing it. Spend a month trying to eliminate mains hum. Then when it is finally cooking nicely, it will sound better than anything you have ever heard in your life. Music of the angels.
It's all psychological of course. But hey, it works
(You will of course have crafted your own speaker cabinets and cross-overs to go with this)
True. Once the distortion figure gets to about 0.1% or less the vast majority if not all listeners cannot tell the difference between a tube amp or solid state one.
The same has been shown to be true for the other components of high fidelity audio systems in numerous tests over the years.
Yeah, this guy cut a few corners....
"There are DSP (digital signal processor) systems that claim to be able to emulate the sound of any amplifier - valve or transistor, and you won't find many valves in the DSP. By one means or another, the dynamic distortion and signal compression abilities of a valve amp can be duplicated very well, and without resorting to a token valve in the circuit. Valve amp distortion comes with the valves, but with opamps and transistors (or DSP) you can adjust the circuitry to be as clean or dirty as you like - user selectable. "
Not just the amplifiers (DSP) but also speakers, microphones, FX, etc. Truly amazing stuff.
http://www.rolandus.com/products/vg-8ex/
That thing made my $49 dollar Strat knockoff sound like a $1000 Martin. I like DSP where it will clean up signals, not distort them. But that's me.
I also read up on class a, b, ab, c, and d amplifiers. I've made simple a class transistor amplifiers, but none of the others. Class d's are pretty slick and probably in most of the modern stuff I've been working with.
All of this almost makes me want to build a radio. Also, more tube dark sorcery:
Just like this one: http://www.ebay.com/itm/Elements-of-Radio-by-A-Marcus-1943-3rd-printing-Tube-Radio-Guide-/262074608473
I hear ya. My personal favourite is the Digitech GNX4 but with a 3rd party library, "mfxsupermodels".
I have to believe that Digitech made a huge commercial mistake with this amazing device. They included WAY too much for the price.
Their newer models don't come close.
I have been buying my "backup" units off ebay for a very low price. I go for the ones that have only had bedroom use. The last one that I purchased, the guy threw in a Roland VGA-3 amplifier which is a keeper.